BRIEF COMMUNICATION

aB-crystallin/HspB5 regulates endothelial–leukocyte interactionsby enhancing NF-jB-induced up-regulation of adhesion moleculesICAM-1, VCAM-1 and E-selectin

Lothar C. Dieterich • Hua Huang • Sara Massena •

Nikola Golenhofen • Mia Phillipson •

Anna Dimberg

Received: 3 May 2013 / Accepted: 13 July 2013 / Published online: 9 August 2013

� The Author(s) 2013. This article is published with open access at Springerlink.com

Abstract aB-crystallin is a small heat shock protein,

which has pro-angiogenic properties by increasing survival

of endothelial cells and secretion of vascular endothelial

growth factor A. Here we demonstrate an additional role of

aB-crystallin in regulating vascular function, through

enhancing tumor necrosis factor a (TNF-a) induced

expression of endothelial adhesion molecules involved in

leukocyte recruitment. Ectopic expression of aB-crystallin

in endothelial cells increases the level of E-selectin

expression in response to TNF-a, and enhances leukocyte–

endothelial interaction in vitro. Conversely, TNF-a-

induced expression of intercellular adhesion molecule 1,

vascular cell adhesion molecule 1 and E-selectin is mark-

edly inhibited in endothelial cells isolated from aB-crys-

tallin-deficient mice. This is associated with elevated levels

of IjB in aB-crystallin deficient cells and incomplete

degradation upon TNF-a stimulation. Consistent with this,

endothelial adhesion molecule expression is reduced in

inflamed vessels of aB-crystallin deficient mice, and leu-

kocyte rolling velocity is increased. Our data identify aB-

crystallin as a new regulator of leukocyte recruitment, by

enhancing pro-inflammatory nuclear factor j B-signaling

and endothelial adhesion molecule expression during

endothelial activation.

Keywords aB-crystallin � Chaperone � ICAM-1 �VCAM-1 � E-selectin � NF-jB

Introduction

aB-crystallin (HspB5) is a member of the small heat shock

protein family, which is ubiquitously expressed in various

cell types and tissues, including endothelial cells, and which

can be further induced in response to stress [1, 2]. Through

its chaperone function, aB-crystallin affects diverse cellular

processes including cytoskeletal rearrangement, production

of reactive oxygen species, proliferation and cell survival

(reviewed in [3, 4]). Recently, aB-crystallin has emerged as

an important regulator of angiogenesis. We have shown that

aB-crystallin is up-regulated in endothelial cells during

angiogenesis and protects endothelial cells from apoptosis

by inhibiting activation of pro-caspase-3 [1]. Consequently,

tumor angiogenesis is significantly reduced in cryab -/-

mice, the vessels are hyper-permeable and frequently show

signs of endothelial apoptosis. Additionally, aB-crystallin

has also been implicated in regulation of physiological and

pathological angiogenesis by stabilizing and promoting

secretion of vascular endothelial growth factor A (VEGF-

A) [5, 6].

Lothar C. Dieterich and Hua Huang contributed equally to this work.

Electronic supplementary material The online version of thisarticle (doi:10.1007/s10456-013-9367-4) contains supplementarymaterial, which is available to authorized users.

L. C. Dieterich � H. Huang � A. Dimberg (&)

The Rudbeck Laboratory, Department of Immunology, Genetics

and Pathology, Uppsala University, 751 85 Uppsala, Sweden

e-mail: Anna.Dimberg@igp.uu.se

Present Address:

L. C. Dieterich

Institute of Pharmaceutical Sciences, Swiss Federal Institute

of Technology (ETH) Zurich, 8093 Zurich, Switzerland

S. Massena � M. Phillipson

Department of Medical Cell Biology, Uppsala University,

751 23 Uppsala, Sweden

N. Golenhofen

Institute of Anatomy and Cell Biology, University of Ulm,

Albert-Einstein-Allee 11, 89081 Ulm, Germany

123

Angiogenesis (2013) 16:975–983

DOI 10.1007/s10456-013-9367-4

Interestingly, we have recently demonstrated a role for

aB-crystallin in regulating chronic inflammatory processes,

suggesting an additional path through which aB-crystallin

expression may affect angiogenesis during pathological

conditions. We found that aB-crystallin is expressed in

spleen derived, immature CD11b?Gr-1? myeloid cells

(IMCs), whereas expression in other leukocyte subsets,

including mature granulocytes, is negligible [7]. We found

that accumulation of IMCs is increased in tumor bearing

cryab -/- mice due to an IMC-intrinsic effect of aB-crys-

tallin. However, a potential contribution of endothelial aB-

crystallin to this phenotype has not been investigated. A

central step in the inflammatory process is the activation of

endothelial cells that mediate recruitment of leukocytes

from the blood stream into the inflamed tissue. Pro-

inflammatory cytokines such as tumor necrosis factor a(TNF-a) are produced by tissue resident immune cells,

inducing activation of local endothelial cells through the

nuclear factor j B (NF-jB) pathway, which in endothelial

cells regulates the expression of various chemokines and

adhesion molecules such as E-selectin, intercellular adhe-

sion molecule 1 (ICAM-1), and vascular cell adhesion

molecule 1 (VCAM-1) [8]. Together, these proteins mediate

the capture and adhesion of leukocytes in the blood stream

in a well-known multistep process (reviewed in [9, 10]).

In this report, we have employed in vitro models of

endothelial activation and intravital microscopy of

inflamed vessels to investigate the role of aB-crystallin in

regulating inflammatory activation of endothelial cells. Our

data demonstrates that aB-crystallin has an important

function during this process by enhancing TNF-a-induced

activation of NF-jB-signalling and downstream activation

of endothelial adhesion molecules.

Materials and methods

Mice

Mice deficient for aB-crystallin and HspB2 (cryab -/-)

were described previously [11]. 129S6/SvEvTac wild type

mice were obtained from Taconic M&B (Bomholt, Den-

mark). All animal work was performed according to the

guidelines for animal experimentation and welfare pro-

vided by Uppsala University and approved by a regional

ethics committee.

Cells

Human embryonic kidney (HEK) 293 T cells were

obtained from ATCC (Manassas, VA) and maintained in

DMEM (Life Technologies, Carlsbad, CA) ? 10 % FCS

(Sigma-Aldrich, St. Louis, MO).

Jurkat cells were from ATCC and were cultured in

RPMI medium (Life Technologies) ? sodium pyruvate

(Life Technologies) and 10 % FCS.

Human Umbilical Vein Endothelial Cells (HUVEC) and

Human Dermal Microvascular Cells (HDMEC) were from

3H Biomedicals (Uppsala, Sweden) and were maintained

on gelatinized culture dishes in endothelial basal medium

(EBM-MV2, PromoCell, Heidelberg, Germany) with full

supplements. Primary cells were used until passage 10.

MyEnd cells have been described previously [2] and

were cultured in DMEM ? 10 % FCS.

Production of lentivirus and transduction of HUVEC

For synthesis of the lentivirus vector pgk:cryab, a full

length clone of human cryab cDNA (NM_001885.1, Ori-

gene, Rockville, MD) was cloned into a modified lentiviral

pgk vector [12] containing an internal ribosomal entry site

(IRES) and eGFP as selection marker. For production of

lentivirus, subconfluent HEK 293 T cells were transfected

with the pgk:cryab construct and the third generation

packaging vectors pMDLg/pRPE, pRSV-rev and pMD2.G

[13] by calcium phosphate precipitation. The medium was

changed the next day and supernatant containing virus

particles was collected 48 h later, filtered (0.45 lm pore

size) and stored at -80 �C until usage. The concentration

of infectious particles was determined by transducing HEK

293 T cells with serious dilutions of supernatants of the

virus pgk:cryab or a control virus (pgk:ev) containing no

cDNA and measurement of GFP expression 72 h later on a

LSRII FACS instrument (BD, Franklin Lakes, NJ). HU-

VEC cells at passage 4 were seeded on gelatinized culture

dishes 1 h prior to transduction with lentivirus containing

supernatants at an MOI of 1. Polybrene (Sigma-Aldrich)

was added to a final concentration of 5 lg/ml to improve

transduction efficiency. Medium was exchanged after 12 h

and cells were cultured for an additional 3 passages before

sorting of GFP? cells on a FACSVantage SE (BD). GFP

expression was tested routinely by FACS during further

culturing of the cells.

In vitro stimulation of endothelial cells

Endothelial cells were rinsed with PBS and treated with

recombinant mouse TNF-a (20 lg/ml, BioVision, Moun-

tain View, CA) in EBM MV2 ? 1 % FCS (for HUVEC)

and DMEM (for MyEnd) for the indicated time periods.

RNA extraction, cDNA synthesis and quantitative

real-time PCR (qPCR)

RNA from adherently growing cells and cremaster muscles

was extracted using the RNeasy Mini Kit (Qiagen, Hilden,

976 Angiogenesis (2013) 16:975–983

123

Germany) or Micro Kit, respectively, with on-column

DNase digestion, according to the manufacturer’s instruc-

tions. Between 500 ng and 1 lg of RNA were used for

reverse transcription using random primers and SuperScript

III reverse transcriptase (Life Technologies) according to the

manufacturer’s instructions. Quantitative real-time PCR was

done using Sybr-Green (Life Technologies), 0.25 lM for-

ward and reverse primer and 0.5 ll cDNA per reaction, with

human/mouse hprt serving as internal control. For primer

sequences, please refer to table S1 in the supplement.

All reactions were performed in triplicates on an

MX3005 instrument (Stratagene, Cedar Creek, TX). For

gene expression analysis, relative expression values were

calculated according to the formula: relative expression

gene x = 2^ - (Ct gene x - Ct internal reference) and the mean

expression and standard deviation for each triplicate was

calculated.

Western blot

Cells were washed with PBS and lysed with 19 LDS

sample buffer and 19 sample reducing agent (Life Tech-

nologies). After homogenization by vigorous pipetting and

incubation at 70 �C for 10 min, samples were separated on

NuPage 4-12 % Bis–Tris Gels using MOPS buffer (Life

Technologies) and transferred to Hybond-C extra (GE

Healthcare, Chalfont St. Giles, UK) according to the

manufacturer’s protocol. Membranes were blocked in 5 %

non-fat dry milk in TBS ? 0.01 % Tween (blocking

solution) for 1 h. Primary antibodies diluted in blocking

solution were incubated over night at 4 �C. Membranes

were washed in TBS-T and incubated with HRP conjugated

secondary antibodies. Membranes were washed several

times in TBS-T before detection using ECL Prime sub-

strate or CCD camera detection using Bio-Rad Chemi-

DocTM MP Imaging System (Bio-Rad, Hercules, CA).

Primary antibodies used were anti-aB-crystallin (Clone

1B6.1-3G4, Enzo LifeSciences, Farmingdale, NY), anti-

Actin (sc-1615, Santa Cruz Biotech, Santa Cruz, CA), anti-

b-catenin (610153, BD Biosciences), and anti-IjB (sc-371,

Santa Cruz). Secondary antibodies were donkey anti-

mouse-HRP (GE Healthcare), donkey anti-rabbit-HRP (GE

Healthcare) and mouse anti-goat-HRP (Clone GT-34,

Sigma-Aldrich).

Nuclear translocation of p-p65

Human Umbilical Vein Endothelial Cells grown on cover-

slips were starved and activated as described above. At the

indicated timepoints, cells were washed with TBS, fixed in

zinc fix (0.1 M Tris–HCl, pH 7.5, 3 mM calcium acetate,

23 mM zinc acetate, and 37 mM zinc chiloride) with 0.2 %

Triton 9-100 for 20 min at RT, washed, and blocked in 3 %

FBS/TBS for 1 h at RT. Cells were stained with rabbit anti-

pSer536-p65 (Cell Signaling, Danvers, MA), washed 3

times with TBS, incubated with donkey anti-rabbit-

Alexa488 (Life Technologies), and counterstained with

Hoechst33342 (2 lg/ml) before mounting. Images were

taken on a Nikon Eclipse fluorescence microscope and

analyzed using ImageJ (NIH, Bethesda, MD).

FACS analysis

Endothelial cells were washed with PBS ? 1 mM EDTA

and gently detached using 0.01 % Trypsin in PBS ? 1 mM

EDTA. FACS buffer (1 % FCS, 0.02 % NaN3 in PBS) was

added immediately once the cells became detached. Cells

were incubated with primary antibodies diluted in FACS

buffer for 1 h at 4 �C, washed with FACS buffer, and

subsequently incubated with secondary antibodies for an

additional 30 min. Directly before FACS analysis, cells

were washed and resuspended in FACS buffer, and DAPI

or PI (Sigma-Aldrich) were added to discriminate living

and dead cells. Samples were analyzed on a LSRII

cytometer (BD). The following primary antibodies were

used (at 2 lg/ml): mouse control IgG1 (BD), mouse anti-

human ICAM-1, mouse anti-human E-selectin, goat anti-

mouse ICAM-1, goat anti-mouse VCAM-1 (all R&D

Systems, Minneapolis, MN), mouse anti-human VCAM-1

(eBioscience, San Diego, CA) and PE labeled rat anti-

mouse E-selectin (BD). The following secondary antibod-

ies were used (10 lg/ml): goat anti-mouse IgG conjugated

to Alexa488, goat anti-mouse IgG conjugated to Alexa555,

and donkey anti-goat IgG conjugated to Alexa488 (all Life

Technologies).

Adhesion assay

Jurkat cells were labeled with the PKH26 red fluorescent

cell linker kit (Sigma-Aldrich) according the manufacturer’s

instructions and co-incubated with HUVEC monolayers in

24 well plates for 15 min at 37 �C on an orbital shaker to

mimic flow. Subsequently, monolayers were washed 4 times

with PBS and fixed in 4 % PFA before microscopic exam-

ination on an LSM700 inverted confocal microscope (Carl

Zeiss, Jena, Germany). The number of adherent Jurkat cells

per well was determined using ImageJ (NIH).

Intravital microscopy

Male cryab -/- and wild type mice were anaesthetized with

isoflurane (Abbott Laboratories, Abbott, IL) and the cre-

master muscle was exposed for intravital microscopic

observation of leukocytes as previously described [14]. An

intravital microscope (Ortholux II, Leica Microsystems,

Wetzlar, Germany) equipped with a 250/0.6 W long distance

Angiogenesis (2013) 16:975–983 977

123

water dipping objective (Leica Micorsystems) and a C3077

digital camera (Hamamatsu Photonics, Hamamatsu City,

Japan) and an intravital microscope (DM5000 B, Leica

Micorsystems) equipped with a 209/0.5 W long distance

water dipping objective and an ORCA C10600 digital

camera (Hamamatsu Photonics) were used. For induction of

inflammation, recombinant mouse TNF-a (500 lg/kg body

weight, R&D Systems, in sterile saline) was injected intra-

scrotally 2.5 h prior to the surgical procedure. Single

unbranched venules (35–50 lm in diameter) were chosen for

observation throughout the experiment. 5 min long

sequences were recorded for offline playback analysis at 3.5,

4.0 and 4.5 h after TNF-a administration. The flux of rolling

cells was assumed as the average number of rolling leuko-

cytes per min passing. Rolling velocity was measured for the

first 10 neutrophils entering the field of view at each time

point. Leukocytes were considered adherent if they remained

stationary for at least 30 s, and total leukocyte adhesion was

quantified as the number of adherent cells within a 100 lm

length of venule. Leukocyte emigration was defined as the

number of cells in the extravascular space within the field of

view at the end of each 5 min sequence. Only cells adjacent

to and clearly outside the venule were counted as emigrated.

Statistical analysis

Statistical analysis was done using GraphPad Prism 5.0

(GraphPad Inc., La Jolla, CA). For comparison of two

groups, student’s unpaired t test was used. For comparison

of multiple groups, two-way ANOVA with Bonferroni

post-test was used. A p value \0.05 was considered to be

statistically significant.

Results

Ectopic expression of aB-crystallin in HUVEC

increases TNF-a induced expression of E-selectin

To analyze the role of aB-crystallin in endothelial activa-

tion, we initially used human umbilical vein endothelial

cells that essentially do not express aB-crystallin under

basal conditions. Using lentiviral transduction we created

HUVEC stably expressing aB-crystallin under control of

the pgk promotor. In these cells (pgk:cryab), aB-crystallin

is readily detectable by western blot, while no aB-crystallin

is detected in cells transduced with an empty control vector

(pgk:ev) (Fig. 1a). Overexpression of aB-crystallin in

HUVEC had no significant effects on cell proliferation (Fig

S1a) To determine if aB-crystallin expression affects

expression of endothelial adhesion molecules, we treated

pgk:cryab and pgk:ev HUVEC with TNF-a and analyzed

surface expression of ICAM-1, VCAM-1 and E-selectin by

FACS. Notably, we found a clear increase in mean fluo-

rescent intensity of surface E-selectin staining on

pgk:cryab compared to pgk:ev HUVEC after 24 h but not

5 h of TNF-a stimulation, while surface levels of ICAM-1

and VCAM-1 were similar at both time points (Fig. 1b, c

and Fig S1b-c). Accordingly, the percentage of E-selectin

positive cells was significantly higher after 24 h of TNF-atreatment in pgk:cryab HUVEC (58 %) as compared to

pgk:ev HUVEC (44 %). Western blot analysis revealed

increased levels of E-selectin in pgk:cryab HUVEC

lysates, indicating that the aB-crystallin-induced increase

in surface expression was associated with increased protein

levels (Fig S1d-e). Consistent with this, gene expression

analysis by quantitative real-time PCR (qPCR) revealed an

increase in TNF-a-induced mRNA expression of E-selectin

after 24 h of stimulation in pgk:cryab HUVEC as com-

pared to control cells (Fig. 1d). Collectively, this data

suggests that ectopic expression of aB-crystallin enhances

E-selectin levels mainly through increasing TNF-a-induced

transcriptional activation of gene expression.

Ectopic expression of aB-crystallin in HUVEC leads

to enhanced leukocyte adhesion to endothelium

To assess the functional impact of aB-crystallin associated

increase in surface E-selectin in HUVEC, we analyzed leu-

kocyte adhesion to endothelial cells in vitro. To measure

leukocyte adhesion, monolayers of pgk:cryab and pgk:ev

HUVEC were stimulated with TNF-a for 24 h. Subsequently

we added fluorescently labeled Jurkat cells, a lymphoblastic

cell line which expresses receptors for ICAM-1, VCAM-1

and ligands for E-selectin, and incubated them for 15 min on

an orbital shaker to simulate flow. Firm adhesion of Jurkat

cells was determined by microscopy after extensive wash-

ing. Under these conditions, an increased number of Jurkat

cells adhered to pgk:cryab HUVEC as compared to pgk:ev

HUVEC (Fig. 1e). This suggests that aB-crystallin enhances

capture of leukocytes, consistent with its role in up-regu-

lating E-selectin.

Decreased TNF-a-induced expression of endothelial

adhesion molecules in aB-crystallin-deficient

microvascular endothelial cells

To investigate the role of endogenous aB-crystallin in

inflammatory activation of endothelial cells, we used

endothelial cell lines established from myocardial micro-

vascular endothelial cells isolated from 129S6 wild type

mice (MyEnd wt) or aB-crystallin-deficient (cryab -/-) mice

(MyEnd cryab -/-) [2]. As expected, MyEnd wt endothelial

cells expressed robust levels of aB-crystallin while no aB-

crystallin could be detected in MyEnd cryab -/- cells

(Fig. 2a). To analyze if aB-crystallin deficiency affected

978 Angiogenesis (2013) 16:975–983

123

endothelial activation, cells were treated with TNF-a and

mRNA levels of E-selectin, ICAM-1 and VCAM-1 were

measured by qPCR. We noticed a striking *50 % reduction

in E-selectin mRNA levels in MyEnd cryab -/- cells as

compared to MyEnd wt after 3 h of TNF-a stimulation, and

differences in expression were maintained up to 24 h

(Fig. 2b). Interestingly, aB-crystallin deficiency in MyEnd

cells also decreased TNF-a-induced mRNA expression of

ICAM-1 and VCAM-1 (Fig. 2c, d). In line with this, surface

expression of E-selectin, ICAM-1 and VCAM-1 were

reduced in cryab -/- cells after 5 h of TNF-a treatment (Fig

S2a-c). We conclude that aB-crystallin deficiency is asso-

ciated with reduced TNF-a-induced mRNA expression of

E-selectin, ICAM-1 and VCAM-1, while ectopic expression

of aB-crystallin only affects E-selectin expression in our

experimental set-up.

pgk:cryabpgk:ev

B-Cry

Actin

++-

-

a

- 5 h

b

TNF-

TNF-

d

- 3 h 6 h 12 h 24 h - +TNF-

e Adhesion Assay

0.0

0.5

1.0

1.5

2.0

2.5

pgk:cry

pgk:ev

Rel

ativ

e ad

hes

ion

Rel

. fl.

inte

nsi

ty

0.0

0.5

1.0

1.5

2.0

Rel

. fl.

inte

nsi

ty

Surface E-selectin

Fo

ld e

xpre

ssio

n

- 24 h

*

0.0

0.5

1.0

1.5

2.0

2.5

0.0

0.5

1.0

1.5

*

E-selectin mRNA expression

*

- 24 hTNF-

pgk:

crya

bpg

k:ev

E-selectin

c

Fig. 1 Ectopic expression of aB-crystallin increases E-selectin levels

in HUVEC. a Representative Western Blot for aB-crystallin in

HUVEC transduced with a lentiviral vector coding for full-length

human aB-crystallin (pgk:cryab) or an empty control vector (pgk:ev).

Actin served as loading control. b Representative FACS plots of

E-selectin staining on HUVEC transduced with pgk:cryab and pgk:ev

after treatment with TNF-a for 24 h. c Quantification of surface

expression of E-selectin in HUVEC transduced with pgk:cryab (white

bars) or empty vector controls (black bars). Cells were treated with

TNF-a for 5 h (left panel) or 24 h (right panel) d HUVEC transduced

with pgk:cryab (white bars) or empty vector controls (black bars)

were stimulated for 3, 6, 12 and 24 h with TNF-a and expression of

E-selectin relative to hprt was determined by qPCR (Bars represent

mean ± SD fold expression compared to pgk:ev after 3 h of TNF-a(normalized data from 3 independent experiments), * = p \ 0.05).

e HUVEC transduced with pgk:cryab (white bars) or empty vector

controls (black bars) were grown to confluency and stimulated with

TNF-a for 24 h before co-incubation with Jurkat cells for 15 min.

Firmly adherent cells were microscopically quantified. (Bars repre-

sent mean ± SD (normalized data from 3 individual independent),

* = p \ 0.05)

Angiogenesis (2013) 16:975–983 979

123

TNF-a-induced NF-jB activation is reduced

in the absence of aB-crystallin

Since aB-crystallin has been implicated in both positive and

negative regulation of the NF-jB-pathway in different cell

types [15, 16], we investigated if aB-crystallin modulates

NF-jB activation in endothelial cells. Through western blot

analysis of TNF-a-stimulated MyEnd cryab -/- and MyEnd

wt cells, we found elevated levels and reduced degradation of

IjB in aB-crystallin-deficient endothelial cells (Fig. 3).

Furthermore, nuclear translocation of phospho-NF-jB

p65 was impaired in cryab -/- MyEnd cells (Fig S2c,d). This

demonstrates that absence of aB-crystallin leads to increased

IjB expression and inhibition of NF-jB mediated activation

of endothelial cells, resulting in reduced expression of NF-

jB target genes ICAM-1, VCAM-1 and E-selectin in MyEnd

cryab -/- cells.

Leukocyte–endothelial interactions are altered in aB-

crystallin deficient (cryab -/-) mice

To test whether aB-crystallin dependent modulation of

endothelial adhesion molecules affects leukocyte–endo-

thelial interactions in vivo, we used intravital microscopy to

study leukocyte rolling, adhesion and emigration in

inflamed cremaster venules in cryab -/- and wild type mice.

A single intrascrotal injection of TNF-a was used to induce

endothelial activation. In line with reduced expression of

E-selectin, ICAM-1 and VCAM-1, we found that leukocyte

rolling velocity was significantly higher in cryab -/- mice

than in wild type mice, 4 and 4.5 h after injection of TNF-a(Fig. 4a). We also observed a higher total number of rolling

leukocytes in cryab -/- mice (rolling flux, Fig. 4b), while the

number of firmly adherent leukocytes and emigrated leu-

kocytes was not significantly altered (Fig S3 a-b). This was

E-selectin mRNA expression

Fo

ld e

xpre

ssio

n

0

10

20

30

40MyEnd wt

MyEnd cryab -/-

*

0

5

10

15

* *

0

20

40

60

80

* *

*

*

TNF- − − + + − − + + − − + + − − + +

a

c

B-Cry

Actin

b

d

MyEnd

wt cryab -/-

3h 6h 18h 24h 3h 6h 18h 24h

3h 6h 18h 24h

Fo

ld e

xpre

ssio

n

Fo

ld e

xpre

ssio

n

TNF- − − + + − − + + − − + + − − + + TNF- − − + + − − + + − − + + − − + +

VCAM-1 mRNA expressionICAM-1 mRNA expression

Fig. 2 TNF-a-induced

endothelial activation is reduced

in endothelial cells derived from

cryab -/- mice. a Representative

western blot showing aB-

crystallin expression in MyEnd

wild type and MyEnd cryab -/-

cells. b, c, d MyEnd wild type

and MyEnd cryab -/- cells were

treated with TNF-a for 3, 6, 18

and 24 h and expression of

E-selectin (b), ICAM-1 (c) and

VCAM-1 (d) relative to hprt

was determined by qPCR. (Bars

represent mean ± SD

(normalized data from 3

independent experiments),

* = p \ 0.05)

wt ko wt ko wt ko wt ko

15 min 30 min

wt ko wt ko wt ko wt ko

45 min 60 min

I B

-catenin

TNF- - - + + - - + + - - + + - - + +

Fig. 3 TNF-a-induced activation of NF-jB is reduced in the absence

of aB-crystallin. MyEnd wt and MyEnd cryab -/- cells were treated

with TNF-a for 15, 30, 45 min and 1 h, protein lysates were prepared

and expression of IjB was determined by western blot. One

representative blot of three independent experiments is shown

980 Angiogenesis (2013) 16:975–983

123

associated with decreased mRNA levels of ICAM-1 and

VCAM-1 in TNF-a treated cryab -/- cremaster muscles as

analyzed by qPCR (Fig. 4c, d), while CD31 mRNA levels

were unchanged (Fig S3c). Vascular expression of ICAM-1

and VCAM-1 was confirmed by immunofluorescence

staining (Fig S3d). A trend towards lower E-selectin mRNA

expression in cryab -/- cremaster muscles was also noted,

but did not reach statistical significance (Fig. 4e). Taken

together, these results are consistent with a role for aB-

crystallin in regulating endothelial–leukocyte interactions

by increasing adhesion molecule expression on activated

endothelium in vivo.

Discussion

aB-crystallin has been attributed diverse cellular functions,

including cytoskeletal stabilization, regulation of reactive

oxygen species, and potent anti-apoptotic activity. Various

types of endothelial cells, including human dermal

microvascular endothelial cells (HDMEC) and bovine

capillary endothelial cells (BCE), express aB-crystallin in

culture, while expression is not detectable in e.g. HUVEC.

aB-crystallin expression is up-regulated in tumor associ-

ated blood vessels [1, 17], but the vascular expression

pattern of aB-crystallin in different organs, vascular beds,

and different types of pathologies has not been thoroughly

investigated. We have previously shown that aB-crystallin

is up-regulated during VEGF-A-induced tubular morpho-

genesis and promotes angiogenesis by inhibiting caspase-3

activation in endothelial cells, thus increasing cell survival

[1]. Subsequently, Kase et al. demonstrated an important

role of aB-crystallin in increasing stability and secretion of

VEGF during physiological angiogenesis [5]. Here, we

describe an additional, formerly unappreciated function of

aB-crystallin in regulating adhesion molecules during

endothelial activation.

In vitro, we found that aB-crystallin enhances TNF-ainduced NF-jB activation and expression of adhesion

molecules in endothelial cells. Ectopic expression of aB-

Time (hours post injection)

wild typecryab -/-

Cel

ls /

min

Time (hours post injection)

µm

/ se

c

ba

0

2

4

6

8

10R

el. m

RN

A/h

prt

Rel

. mR

NA

/hp

rt

Rel

. mR

NA

/hp

rtVCAM-1 mRNA expression

0.0

0.5

1.0

1.5

2.0

ICAM-1 mRNA expression

E-selectin mRNA expression

0.0

0.1

0.2

0.3

dc

e

xulfgnilloRyticolevgnilloR

0

10

20

30

40

50

0

50

100

150

200

wild type cryab epytdliw-/- cryab -/-

wild type cryab -/-

* * * *

* *

Fig. 4 In vivo leukocyte–

endothelial interactions are

altered in cryab -/- mice,

associated with decreased

expression of endothelial

adhesion molecules. Intravital

microscopy was used to analyze

leukocyte–endothelial

interactions in inflamed venules

of the cremaster muscle of wild

type (white bars) and cryab -/-

mice (black bars). Analysis of

rolling velocity (a) and rolling

flux (b) 3.5, 4, and 4.5 h after

intrascrotal injection of TNF-a(n = 5, mean ± SD,

*p \ 0.05). qPCR analysis of

mRNA expression of ICAM-1

(c), VCAM-1 (d) and E-selectin

(e) relative to hprt in mouse

cremaster muscles harvested

4.5 h after intrascrotal injection

of TNF-a (n = 4, mean ± SD,

*p \ 0.05)

Angiogenesis (2013) 16:975–983 981

123

crystallin in HUVEC resulted in increased E-selectin

expression and leukocyte adhesion in response to TNF-a.

Both protein and mRNA levels of E-selectin were

increased, indicating that aB-crystallin regulates E-selectin

expression on the transcriptional level. Correspondingly,

TNF-a induced expression of adhesion molecules ICAM-1,

VCAM-1, and E-selectin was reduced in aB-crystallin

deficient endothelial cells. In line with these findings, a

decrease in TNF-a induced mRNA expression of ICAM-1,

VCAM-1 and E-selectin was noted in aB-crystallin-defi-

cient mice. aB-crystallin has been suggested to affect the

NF-jB signaling pathway, both positively and negatively,

in different systems [15, 16]. Notably, Adhikari et al. [15]

recently showed that aB-crystallin expression increases

IKKb activity, thereby promoting phosphorylation and

degradation of IjB in response to TNF-a and, corre-

spondingly, activation of NF-jB in the myoblastic cell line

C2C12. In agreement to this, we found that IjB levels were

increased in aB-crystallin-deficient endothelial cells under

basal conditions, and that TNF-a induced degradation of

IjB and nuclear accumulation of phospho-NF-jB p65 was

inhibited if aB-crystallin was absent. This indicates that

aB-crystallin exerts a similar function in endothelial cells

as in myoblasts, i.e. to regulate NF-jB activation by pro-

moting degradation of IjB, possibly by interacting with

and stimulating IKKb. A difference in IjB degradation

however could not be observed in pgk:cryab transduced

HUVECs (data not shown), suggesting that aB-crystallin

may have additional effects on E-selectin expression.

Using intravital microscopy, we found that leukocyte

rolling velocity in TNF-a activated venules is strikingly

higher in aB-crystallin deficient (cryab -/-) mice. This

finding is in line with various studies using genetic mouse

models demonstrating that deficiency in ICAM-1,

E-selectin or E-selectin ligands results in higher leukocyte

rolling speed in vivo [18–20]. We also observed a higher

number of rolling leukocytes (flux) in cryab -/- mice,

indicating a lower grade of activation. However, the

numbers of adherent and emigrated leukocytes were not

significantly different between groups due to great varia-

tions between individuals. Furthermore, we have shown

previously that unchallenged cryab -/- mice have normal

numbers of circulating leukocytes, indicating that the

increased leukocyte flux is not due to a higher leukocyte

number in these mice [7]. Possibly, compensatory mecha-

nisms, for example an increased surface exposure of

P-selectin or other endothelial adhesion molecules, might

also be involved. Nevertheless, lack of aB-crystallin is

clearly associated with altered leukocyte–endothelial

interactions in inflamed venules in vivo. This suggests that

endothelial aB-crystallin may act as a positive regulator of

leukocyte recruitment in inflammatory conditions, partic-

ularly where its expression is induced by e.g. angiogenic or

stress-related signals. Further studies are warranted to test

this hypothesis in vivo.

Acknowledgments We would like to thank Charlotte Wikner for

technical support, Sonia Tugues and Laurens van Meeteren for help

with production of lentiviral vectors, and Eric F. Wawrousek for kind

provision of cryab -/- mice. This work was supported by Grants from

the Swedish Cancer Society, the Swedish Research Council, the

Magnus Bergvall foundation, the Ake Wiberg foundation, Svenska

lakaresallskapet, Clas Groschinsky foundation, Lars Hierta founda-

tion, and Harald and Greta Jeansson foundation. MP and AD were

supported by Junior Researcher positions from the Swedish Research

Council.

Open Access This article is distributed under the terms of the

Creative Commons Attribution License which permits any use, dis-

tribution, and reproduction in any medium, provided the original

author(s) and the source are credited.

References

1. Dimberg A, Rylova S, Dieterich LC, Olsson AK, Schiller P,

Wikner C, Bohman S, Botling J, Lukinius A, Wawrousek EF,

Claesson-Welsh L (2008) AlphaB-crystallin promotes tumor

angiogenesis by increasing vascular survival during tube mor-

phogenesis. Blood 111(4):2015–2023

2. Golenhofen N, Ness W, Wawrousek EF, Drenckhahn D (2002)

Expression and induction of the stress protein alpha-B-crystallin in

vascular endothelial cells. Histochem Cell Biol 117(3):203–209

3. Arrigo AP, Simon S, Gibert B, Kretz-Remy C, Nivon M,

Czekalla A, Guillet D, Moulin M, Diaz-Latoud C, Vicart P (2007)

Hsp27 (HspB1) and alphaB-crystallin (HspB5) as therapeutic

targets. FEBS Lett 581(19):3665–3674

4. Parcellier A, Schmitt E, Brunet M, Hammann A, Solary E,

Garrido C (2005) Small heat shock proteins HSP27 and alphaB-

crystallin: cytoprotective and oncogenic functions. Antioxid

Redox Signal 7(3–4):404–413

5. Kase S, He S, Sonoda S, Kitamura M, Spee C, Wawrousek E,

Ryan SJ, Kannan R, Hinton DR (2010) AlphaB-crystallin regu-

lation of angiogenesis by modulation of VEGF. Blood

115(16):3398–3406. doi:10.1182/blood-2009-01-197095

6. Ghosh JG, Shenoy AK Jr, Clark JI (2007) Interactions between

important regulatory proteins and human alphaB crystallin. Bio-

chemistry 46(21):6308–6317

7. Dieterich LC, Schiller P, Huang H, Wawrousek EF, Loskog A,

Wanders A, Moons L, Dimberg A (2012) AlphaB-Crystallin

regulates expansion of CD11b?Gr-1? immature myeloid cells

during tumor progression. Faseb J. doi:10.1096/fj.12-213017

8. Pober JS, Sessa WC (2007) Evolving functions of endothelial

cells in inflammation. Nat Rev Immunol 7(10):803–815. doi:10.

1038/nri2171

9. Ley K, Laudanna C, Cybulsky MI, Nourshargh S (2007) Getting

to the site of inflammation: the leukocyte adhesion cascade

updated. Nat Rev Immunol 7(9):678–689. doi:10.1038/nri2156

10. Petri B, Phillipson M, Kubes P (2008) The physiology of leu-

kocyte recruitment: an in vivo perspective. J Immunol 180(10):

6439–6446

11. Brady JP, Garland DL, Green DE, Tamm ER, Giblin FJ, Waw-

rousek EF (2001) AlphaB-crystallin in lens development and

muscle integrity: a gene knockout approach. Invest Ophthalmol

Vis Sci 42(12):2924–2934

12. Follenzi A, Ailles LE, Bakovic S, Geuna M, Naldini L (2000)

Gene transfer by lentiviral vectors is limited by nuclear

982 Angiogenesis (2013) 16:975–983

123

translocation and rescued by HIV-1 pol sequences. Nat Genet

25(2):217–222. doi:10.1038/76095

13. Dull T, Zufferey R, Kelly M, Mandel RJ, Nguyen M, Trono D,

Naldini L (1998) A third-generation lentivirus vector with a

conditional packaging system. J Virol 72(11):8463–8471

14. Massena S, Christoffersson G, Hjertstrom E, Zcharia E, Vlo-

davsky I, Ausmees N, Rolny C, Li JP, Phillipson M (2010) A

chemotactic gradient sequestered on endothelial heparan sulfate

induces directional intraluminal crawling of neutrophils. Blood

116(11):1924–1931. doi:10.1182/blood-2010-01-266072

15. Adhikari AS, Singh BN, Rao KS, Rao Ch M (1813) AlphaB-

crystallin, a small heat shock protein, modulates NF-kappaB

activity in a phosphorylation-dependent manner and protects

muscle myoblasts from TNF-alpha induced cytotoxicity. Biochim

Biophys Acta 8:1532–1542. doi:10.1016/j.bbamcr.2011.04.009

16. Ousman SS, Tomooka BH, van Noort JM, Wawrousek EF,

O’Conner K, Hafler DA, Sobel RA, Robinson WH, Steinman L

(2007) Protective and therapeutic role for alphaB-crystallin in

autoimmune demyelination. Nature 448(7152):474–479

17. Ria R, Todoerti K, Berardi S, Coluccia AM, De Luisi A, Mattioli

M, Ronchetti D, Morabito F, Guarini A, Petrucci MT, Dammacco

F, Ribatti D, Neri A, Vacca A (2009) Gene expression profiling

of bone marrow endothelial cells in patients with multiple mye-

loma. Clin Cancer Res 15(17):5369–5378

18. Steeber DA, Campbell MA, Basit A, Ley K, Tedder TF (1998)

Optimal selectin-mediated rolling of leukocytes during inflam-

mation in vivo requires intercellular adhesion molecule-1

expression. Proc Natl Acad Sci USA 95(13):7562–7567

19. Kunkel EJ, Ley K (1996) Distinct phenotype of E-selectin-defi-

cient mice. E-selectin is required for slow leukocyte rolling

in vivo. Circ Res 79(6):1196–1204

20. Hidalgo A, Peired AJ, Wild MK, Vestweber D, Frenette PS

(2007) Complete identification of E-selectin ligands on neutro-

phils reveals distinct functions of PSGL-1, ESL-1, and CD44.

Immunity 26(4):477–489. doi:10.1016/j.immuni.2007.03.011

Angiogenesis (2013) 16:975–983 983

123